maternion/ lfm2.5:350m-q8_0

1,487 3 weeks ago

LFM2.5-350M & LFM2.5-8B-A1B is a fast, memory-efficient hybrid MoE language model. Unsloth Dynamic (UD) quants available.

tools thinking 350m 8b
f7301c3862a8 · 167B
{
"repeat_penalty": 1.05,
"stop": [
"<|startoftext|>",
"<|im_start|>",
"<|im_end|>",
"<|im_start|>user"
],
"temperature": 0.1,
"top_k": 50
}